To set things straight, this is a blog and it seems like I am badmouthing something that I am personally a part of. But I felt the urge to do so after seeing a series of examples that underlined why blogs suck at most times.

I see that people are talking about the demise of traditional publishing (e.g. newspapers) and the rise of the blogosphere lately. There is all these hot topics about social sharing, democratization and diversification of news (and news sources). Suddenly, editors found themselves the target of critics, as if they were responsible for delivering faulty, wrongly-opinionated, low-quality news to their reader for the last decades. Most of the blogs are on a roll, with millions of viewers and north-facing Analytics graphs. But everyday, I feel the decay of quality in these popular blogs a little bit more. Why shouldn’t I? When a writer is incentivized by the number of the posts she writes (along with a set of other metrics such as # of times a story is read/clicked etc.) and when you eliminate the huge constraint of publishing market (finite publishing space, such as # of pages in newspapers or duration for a TV News run), what you get is as follows:

  • Lots of news
  • Low-quality content
  • Speed delivery

There certainly are good things in these bullet points. Therefore, I want to write more about the low-quality issue. I have a good anecdote to illustrate the immense difference between an ordinary (low-quality blog post) and a story you would read in New York Times. The story is very interesting by itself; Silicon Alley Insider talks about the alleged drop in Wired magazine’s advertising pages, and therefore revenues. The important thing for you to know, is that my aim doesn’t have anything to do with proving SAI is wrong. Instead, I will focus on the mistakes in SAI’s reasoning and arguments. That is something that I consider equally problematic for a news source: Wrong reasoning is as bad as wrong conclusions in my opinion.

The argument #1: The issue numbers just 113 pages in total. Wired‘s January issue contained 128 pages; the December issue, 231 pages.

The argument #2: Of those 113 pages, only 31.5 are ad pages. The usual ratio between editorial and advertising hovers around 1:1. 31.5 ad pages is a 27% decline from the January 2009 issue, which itself was a 47% decline from January 2007.

The conclusion: That is miserable.

Before jumping into critics, I want to underline the perspective again. I personally believe Wired is probably in a big problem. But, if I want to support my thoughts in my blog, the data and arguments I present should be at least logically meaningful. That is not the case in the story above. Yet, it is very easy to overlook these mistakes and reach the same conclusion of SAI in this story.

Let’s consider argument #1. February issue has 113 pages, compared to January’s 128 and December’s 231. That looks bad. Or is it? If I ask you to guess the number of greeting cards sold in stores in December, January and February, what would you say? Probably, there would be a pattern similar to Wired’s numbers. I’m not saying that greeting cards and Wired magazine have something in common. But seasonality is a very strong driver in most markets. You can argue that sales (of greeting cards) are not comparable to the number of pages (in Wired). But that is not the point. The point is, when you want to conclude whether a period is better or worse than the past, comparing to the last month may not be a good indicator. If SAI said that February 2009 issue is thinner than February 2008 issue, then that would mean something. But the fact that February issue is thinner than January issue and December issue just presents an invalid argument. If I told you that Wired’s February 2008 had 110 pages, would your opinion change. What if it was 60 pages, or 240 pages? I don’t know the number, but the problem is, the story doesn’t tell neither.

If you look at argument #2, there is a 27% decline from January 2009 issue in the number of ad pages. And January 2009 had a decline of 47% from January 2007! What a convoluted approach. I understand that the best way to illustrate the demise of Wired’s February 2009 issue is to first compare it to January 2009, and then compare the January 2009 issue to January 2007? What happened to historical February data to compare. And why are we travelling back to 2007 when there is a 2008 in the middle? As a critical reader, in this case, you need to be alert for at least two things:

  • Either the writer is so lazy to make a research for the numbers in January/February 2008.
  • Or the writer is handpicking (or cherrypicking) data to make his story more interesting.

In any of the cases, it is a big blame for SAI. And that is basically why I think traditional publishing still rocks and why blogosphere is a terrible source to read opinionated stories. Yet, it is still a great a place to do a number of things such as discovering what is new, and getting some private beta invites.